Single-source SYCL C++ on Xilinx FPGA. Xilinx Research Labs Khronos 2017/11/12 19

Size: px
Start display at page:

Download "Single-source SYCL C++ on Xilinx FPGA. Xilinx Research Labs Khronos 2017/11/12 19"

Transcription

1 Single-source SYCL C++ on Xilinx FPGA Xilinx Research Labs Khronos 2017/11/12 19

2 Khronos standards for heterogeneous systems 3D for the Web - Real-time apps and games in-browser - Efficiently delivering runtime 3D assets Connecting Software to Silicon Vision and Neural Networks - Tracking and odometry - Scene analysis/understanding - Neural Network inferencing Real-time 2D/3D - Virtual and Augmented Reality - Cross-platform gaming and UI - CG Visual Effects Parallel Computation - Machine Learning acceleration - Embedded vision processing - High Performance Computing (HPC) - CAD and Product Design - Safety-critical displays

3 PROMOTER MEMBERS Over 100 members worldwide Any company is welcome to join Copyright Khronos Group 2017

4 Complete example of matrix addition in OpenCL SYCL #include <CL/syclhpp> buffer B { &b[0][0], range { N, M } }; #include <iostream> using namespace cl::sycl; constexpr size_t N = 2; constexpr size_t M = 3; using Matrix = float[n][m]; buffer C { &c[0][0], range { N, M } }; // Enqueue some computation kernel task qsubmit([&](handler& cgh) { // Define the data used/produced auto ka = Aget_access<access::mode::read>(cgh); auto kb = Bget_access<access::mode::read>(cgh); // Compute sum of matrices a and b into c int main() { Matrix a = { { 1, 2, 3 }, { 4, 5, 6 } }; Matrix b = { { 2, 3, 4 }, { 5, 6, 7 } }; auto kc = Cget_access<access::mode::write>(cgh); // Create & call kernel named "mat_add" cghparallel_for<class mat_add>(range { N, M }, [=](id<2> i) { kc[i] = ka[i] + kb[i]; } ); Matrix c; }); // End of our commands for this queue } // End scope, so wait for the buffers to be released {// Create a queue to work on default device queue q; // Wrap some buffers around our data buffer A { &a[0][0], range { N, M } }; Page 4 // Copy back the buffer data with RAII behaviour std::cout << "c[0][2] = " << c[0][2] << std::endl; return 0; }

5 trisycl Open Source SYCL 12/22 Uses C++17 templated classes Used by Khronos to define the SYCL and OpenCL C++ standard Languages are now too complex to be defined without implementing On-going implementation started at AMD and now led by Xilinx OpenMP for host parallelism BoostCompute for OpenCL interaction Prototype of device compiler for Xilinx FPGA Page 5

6 architecture #include <CL/syclhpp> [] C++ SYCL qsubmit([&](auto &cgh) { // The kernel write a, so get a write accessor on it auto A = aget_access<access::mode::write>(cgh); Unmodified host compiler (gcc/clang/vs/icc) For OpenCL interoperability OpenMP CPU executable // Enqueue parallel kernel on a N*M 2D iteration space cghparallel_for<class init_a>({ N, M }, [=] C++ (autosycl index) { A[index] = index[0]*2 + index[1]; }); });} #include <CL/syclhpp> libopenclso Clang/LLVM Host & kernel caller OpenMP CPU executable C++17 & OpenMP & Boost OpenCL interoperability (BoostCompute) SYCL runtime Clang/LLVM device compiler kernelsbin Device Compiler Runtime SPIR 20 de facto Vendor OpenCL device compiler

7 SPIR 20 de facto output with Clang 391 using; ModuleID = 'device_compiler/single_task_vector_add_drtkernelbc' source_filename = "device_compiler/single_task_vector_add_drtcpp" target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-s128" target triple = "spir64" "_ZZZ9test_mainiPPcENK3$_1clERN2cl4sycl7handlerEENKUlvE_clEvexit": ; preds = %forbodyi } ret void declare gxx_personality_v0() ; Function Attrs: noinline norecurse nounwind uwtable define spir_kernel addrspace(1)* %f000val, i32 addrspace(1)* %f010val, i32 addrspace(1)* %f020val) unnamed_addr #0!kernel_arg_addr_space!3!kernel_arg_type!4!kernel_arg_base_type!4!kernel_arg_type_qual!5!kernel_arg_access_qual!6 {!llvmident =!{!0} attributes #0 = { noinline norecurse nounwind uwtable "disable-tail-calls"="false" "less-precisefpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+fxsr,+mmx,+sse,+sse2,+x87" "unsafe-fp-math"="false" "use-soft-float"="false" } entry: br label %forbodyi!openclspirversion =!{!1}!opencloclversion =!{!2} forbodyi: ; preds = %forbodyi, %entry %indvarsivi = phi i64 [ 0, %entry ], [ %indvarsivnexti, %forbodyi ] %arrayidxii = getelementptr inbounds i32, i32 addrspace(1)* %f010val, i64 %indvarsivi %0 = load i32, i32 addrspace(1)* %arrayidxii, align 4,!tbaa!7 %arrayidxi15i = getelementptr inbounds i32, i32 addrspace(1)* %f020val, i64 %indvarsivi %1 = load i32, i32 addrspace(1)* %arrayidxi15i, align 4,!tbaa!7 %addi = add nsw i32 %1, %0 %arrayidxi13i = getelementptr inbounds i32, i32 addrspace(1)* %f000val, i64 %indvarsivi store i32 %addi, i32 addrspace(1)* %arrayidxi13i, align 4,!tbaa!7 %indvarsivnexti = add nuw nsw i64 %indvarsivi, 1 %exitcondi = icmp eq i64 %indvarsivnexti, 300!0 =!{!"clang version 391 "}!1 =!{i32 2, i32 0}!2 =!{i32 1, i32 2}!3 =!{i32 1, i32 1, i32 1}!4 =!{!"int *",!"int *",!"int *"}!5 =!{!"",!"",!""}!6 =!{!"read_write",!"read_write",!"read_write"}!7 =!{!8,!8, i64 0}!8 =!{!"int",!9, i64 0}!9 =!{!"omnipotent char",!10, i64 0}!10 =!{!"Simple C++ TBAA"} br i1 %exitcondi, label %"_ZZZ9test_mainiPPcENK3$_1clERN2cl4sycl7handlerEENKUlvE_clEvexit", label %forbodyi Page 7

8 After Xilinx SDx xocc ingestion Page 8

9 After Xilinx SDx xocc ingestion FPGA layout! Page 9

10 Code execution on real FPGA (device)$ device_compiler/single_task_vector_add_drtkernel_caller binary_size = task::add_prelude task::add_prelude task::add_prelude accessor(accessor &a) : &a = 0x7ffd39395f40 &buffer =0x7ffd39395f50 accessor(accessor &a) : &a = 0x7ffd39395f30 &buffer =0x7ffd39395f60 accessor(accessor &a) : &a = 0x7ffd39395f20 &buffer =0x7ffd39395f70 single_task &f = 0x7ffd39395f50 task::prelude schedule_kernel &k = 0x Setting up _ZN2cl4sycl6detail18instantiate_kernelIZZ9test_mainiPPcENK3$_1clERNS0_7handlerEE3addZZ9test_mainiS4_ENKS5_clES7_EUlvE_EEvT0_ aka TRISYCL_kernel_0 Name device xilinx_adm-pcie-7v3_1ddr_3_0 serialize_accessor_arg index =0, size = 4, arg = 0 serialize_accessor_arg index =1, size = 4, arg = 0x1 serialize_accessor_arg index =2, size = 4, arg = 0x2 **** no errors detected Page 10

11 Page 11

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 SYCL and OpenCL State of the Nation Michael Wong ISOCPP VP Codeplay Vice President of R & D SYCL Working Group Chair Chair C++ Standard SG5, SG14 michael@codeplay.com wongmichael.com Ronan Keryell Xilinx

More information

Single-source SYCL C++ on Xilinx FPGA. Xilinx Research Labs Khronos 2017/11/12 19

Single-source SYCL C++ on Xilinx FPGA. Xilinx Research Labs Khronos 2017/11/12 19 Single-source SYCL C++ on Xilinx FPGA Xilinx Research Labs Khronos booth @SC17 2017/11/12 19 Power wall & speed of light: the final frontier Current physical limits Power consumption Cannot power-on all

More information

Copyright Khronos Group Page 1. Introduction to SYCL. SYCL Tutorial IWOCL

Copyright Khronos Group Page 1. Introduction to SYCL. SYCL Tutorial IWOCL Copyright Khronos Group 2015 - Page 1 Introduction to SYCL SYCL Tutorial IWOCL 2015-05-12 Copyright Khronos Group 2015 - Page 2 Introduction I am - Lee Howes - Senior staff engineer - GPU systems team

More information

Copyright Khronos Group, Page 1 SYCL. SG14, February 2016

Copyright Khronos Group, Page 1 SYCL. SG14, February 2016 Copyright Khronos Group, 2014 - Page 1 SYCL SG14, February 2016 BOARD OF PROMOTERS Over 100 members worldwide any company is welcome to join Copyright Khronos Group 2014 SYCL 1. What is SYCL for and what

More information

OpenCL: History & Future. November 20, 2017

OpenCL: History & Future. November 20, 2017 Mitglied der Helmholtz-Gemeinschaft OpenCL: History & Future November 20, 2017 OpenCL Portable Heterogeneous Computing 2 APIs and 2 kernel languages C Platform Layer API OpenCL C and C++ kernel language

More information

SYCL for OpenCL. in a nutshell. Maria Rovatsou, Codeplay s R&D Product Development Lead & Contributor to SYCL. IWOCL Conference May 2014

SYCL for OpenCL. in a nutshell. Maria Rovatsou, Codeplay s R&D Product Development Lead & Contributor to SYCL. IWOCL Conference May 2014 SYCL for OpenCL in a nutshell Maria Rovatsou, Codeplay s R&D Product Development Lead & Contributor to SYCL! IWOCL Conference May 2014 SYCL for OpenCL in a nutshell SYCL in the OpenCL ecosystem SYCL aims

More information

15-411: LLVM. Jan Hoffmann. Substantial portions courtesy of Deby Katz

15-411: LLVM. Jan Hoffmann. Substantial portions courtesy of Deby Katz 15-411: LLVM Jan Hoffmann Substantial portions courtesy of Deby Katz and Gennady Pekhimenko, Olatunji Ruwase,Chris Lattner, Vikram Adve, and David Koes Carnegie What is LLVM? A collection of modular and

More information

trisycl Open Source C++17 & OpenMP-based OpenCL SYCL prototype Ronan Keryell 05/12/2015 IWOCL 2015 SYCL Tutorial Khronos OpenCL SYCL committee

trisycl Open Source C++17 & OpenMP-based OpenCL SYCL prototype Ronan Keryell 05/12/2015 IWOCL 2015 SYCL Tutorial Khronos OpenCL SYCL committee trisycl Open Source C++17 & OpenMP-based OpenCL SYCL prototype Ronan Keryell Khronos OpenCL SYCL committee 05/12/2015 IWOCL 2015 SYCL Tutorial OpenCL SYCL committee work... Weekly telephone meeting Define

More information

Introducing Parallelism to the Ranges TS

Introducing Parallelism to the Ranges TS Introducing Parallelism to the Ranges TS Gordon Brown, Christopher Di Bella, Michael Haidl, Toomas Remmelg, Ruyman Reyes, Michel Steuwer Distributed & Heterogeneous Programming in C/C++, Oxford, 14/05/2018

More information

Using SYCL as an Implementation Framework for HPX.Compute

Using SYCL as an Implementation Framework for HPX.Compute Using SYCL as an Implementation Framework for HPX.Compute Marcin Copik 1 Hartmut Kaiser 2 1 RWTH Aachen University mcopik@gmail.com 2 Louisiana State University Center for Computation and Technology The

More information

SYCL for OpenCL May15. Copyright Khronos Group Page 1

SYCL for OpenCL May15. Copyright Khronos Group Page 1 SYCL for OpenCL May15 Copyright Khronos Group 2015 - Page 1 Copyright Khronos Group 2015 - Page 2 SYCL for OpenCL - Single-source C++ Pronounced sickle - To go with spear (SPIR) Royalty-free, cross-platform

More information

The Role of Standards in Heterogeneous Programming

The Role of Standards in Heterogeneous Programming The Role of Standards in Heterogeneous Programming Multi-core Challenge Bristol UWE 45 York Place, Edinburgh EH1 3HP June 12th, 2013 Codeplay Software Ltd. Incorporated in 1999 Based in Edinburgh, Scotland

More information

SYCL for OpenCL in a Nutshell

SYCL for OpenCL in a Nutshell SYCL for OpenCL in a Nutshell Luke Iwanski, Games Technology Programmer @ Codeplay! SIGGRAPH Vancouver 2014 1 2 Copyright Khronos Group 2014 SYCL for OpenCL in a nutshell Copyright Khronos Group 2014 Why?

More information

PTX Back-End: GPU Programming with LLVM

PTX Back-End: GPU Programming with LLVM PTX Back-End: GPU Programming with LLVM Justin Holewinski The Ohio State University LLVM Developer's Meeting November 18, 2011 Justin Holewinski (Ohio State) PTX Back-End Nov. 18, 2011 1 / 37 Outline PTX

More information

Creating Formally Verified Components for Layered Assurance with an LLVM-to-ACL2 Translator

Creating Formally Verified Components for Layered Assurance with an LLVM-to-ACL2 Translator Creating Formally Verified Components for Layered Assurance with an LLVM-to-ACL2 Translator Jennifer Davis, David Hardin, Jedidiah McClurg December 2013 Introduction Research objectives: Reduce need to

More information

VisionCPP Mehdi Goli

VisionCPP Mehdi Goli VisionCPP Mehdi Goli Motivation Computer vision Different areas Medical imaging, film industry, industrial manufacturing, weather forecasting, etc. Operations Large size of data Sequence of operations

More information

OpenCL C++ kernel language

OpenCL C++ kernel language Copyright Khronos Group 2016 - Page 1 OpenCL C++ kernel language Vienna April 2016 Adam Stański Bartosz Sochacki Copyright Khronos Group 2016 - Page 2 OpenCL 2.2 OpenCL C++ Open source free compiler https://github.com/khronosgroup/libclcxx

More information

PROGRAMOVÁNÍ V C++ CVIČENÍ. Michal Brabec

PROGRAMOVÁNÍ V C++ CVIČENÍ. Michal Brabec PROGRAMOVÁNÍ V C++ CVIČENÍ Michal Brabec PARALLELISM CATEGORIES CPU? SSE Multiprocessor SIMT - GPU 2 / 17 PARALLELISM V C++ Weak support in the language itself, powerful libraries Many different parallelization

More information

Copyright Khronos Group 2012 Page 1. OpenCL 1.2. August 2012

Copyright Khronos Group 2012 Page 1. OpenCL 1.2. August 2012 Copyright Khronos Group 2012 Page 1 OpenCL 1.2 August 2012 Copyright Khronos Group 2012 Page 2 Khronos - Connecting Software to Silicon Khronos defines open, royalty-free standards to access graphics,

More information

ebpf-based tracing tools under 32 bit architectures

ebpf-based tracing tools under 32 bit architectures Linux Plumbers Conference 2018 ebpf-based tracing tools under 32 bit architectures Maciej Słodczyk Adrian Szyndela Samsung R&D Institute Poland

More information

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 OpenCL State of the Nation Neil Trevett Khronos President NVIDIA Vice President Developer Ecosystem OpenCL Working Group Chair ntrevett@nvidia.com @neilt3d Toronto, May 2017 Copyright Khronos Group 2017

More information

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 OpenCL State of the Nation Neil Trevett Khronos President NVIDIA Vice President Developer Ecosystem OpenCL Working Group Chair ntrevett@nvidia.com @neilt3d Toronto, May 2017 Copyright Khronos Group 2017

More information

OpenCL Press Conference

OpenCL Press Conference Copyright Khronos Group, 2011 - Page 1 OpenCL Press Conference Tokyo, November 2011 Neil Trevett Vice President Mobile Content, NVIDIA President, The Khronos Group Copyright Khronos Group, 2011 - Page

More information

Copyright Khronos Group Page 1. Vulkan Overview. June 2015

Copyright Khronos Group Page 1. Vulkan Overview. June 2015 Copyright Khronos Group 2015 - Page 1 Vulkan Overview June 2015 Copyright Khronos Group 2015 - Page 2 Khronos Connects Software to Silicon Open Consortium creating OPEN STANDARD APIs for hardware acceleration

More information

Advanced OpenMP. Other threading APIs

Advanced OpenMP. Other threading APIs Advanced OpenMP Other threading APIs What s wrong with OpenMP? OpenMP is designed for programs where you want a fixed number of threads, and you always want the threads to be consuming CPU cycles. - cannot

More information

SYCL-based Data Layout Abstractions for CPU+GPU Codes

SYCL-based Data Layout Abstractions for CPU+GPU Codes DHPCC++ 2018 Conference St Catherine's College, Oxford, UK May 14th, 2018 SYCL-based Data Layout Abstractions for CPU+GPU Codes Florian Wende, Matthias Noack, Thomas Steinke Zuse Institute Berlin Setting

More information

The OpenVX Computer Vision and Neural Network Inference

The OpenVX Computer Vision and Neural Network Inference The OpenVX Computer and Neural Network Inference Standard for Portable, Efficient Code Radhakrishna Giduthuri Editor, OpenVX Khronos Group radha.giduthuri@amd.com @RadhaGiduthuri Copyright 2018 Khronos

More information

Parallel STL in today s SYCL Ruymán Reyes

Parallel STL in today s SYCL Ruymán Reyes Parallel STL in today s SYCL Ruymán Reyes ruyman@codeplay.com Codeplay Research 15 th November, 2016 Outline 1 Parallelism TS 2 The SYCL parallel STL 3 Heterogeneous Execution with Parallel STL 4 Conclusions

More information

A Translation Framework for Automatic Translation of Annotated LLVM IR into OpenCL Kernel Function

A Translation Framework for Automatic Translation of Annotated LLVM IR into OpenCL Kernel Function A Translation Framework for Automatic Translation of Annotated LLVM IR into OpenCL Kernel Function Chen-Ting Chang, Yu-Sheng Chen, I-Wei Wu, and Jyh-Jiun Shann Dept. of Computer Science, National Chiao

More information

Visualizing code structure in LLVM

Visualizing code structure in LLVM Institute of Computational Science Visualizing code structure in LLVM Dmitry Mikushin dmitry.mikushin@usi.ch. December 5, 2013 Dmitry Mikushin Visualizing code structure in LLVM December 5, 2013 1 / 14

More information

REFLECTIONS ON TRUSTING BITCODE

REFLECTIONS ON TRUSTING BITCODE ITSECX - @FREDERICJACOBS REFLECTIONS ON TRUSTING BITCODE 2 INTRO ~ WHOIS FREDERIC icepa Signal Both Open-Source Software INTRO 3 Talk based on blog post HTTPS://MEDIUM.COM/@FREDERICJACOBS/WHY-I-M-NOT-ENABLING-BITCODE-F35CD8FBFCC5

More information

Open Standards for Vision and AI Peter McGuinness NNEF WG Chair CEO, Highwai, Inc May 2018

Open Standards for Vision and AI Peter McGuinness NNEF WG Chair CEO, Highwai, Inc May 2018 Copyright Khronos Group 2018 - Page 1 Open Standards for Vision and AI Peter McGuinness NNEF WG Chair CEO, Highwai, Inc peter.mcguinness@gobrach.com May 2018 Khronos Mission E.g. OpenGL ES provides 3D

More information

OpenCL Overview. Shanghai March Neil Trevett Vice President Mobile Content, NVIDIA President, The Khronos Group

OpenCL Overview. Shanghai March Neil Trevett Vice President Mobile Content, NVIDIA President, The Khronos Group Copyright Khronos Group, 2012 - Page 1 OpenCL Overview Shanghai March 2012 Neil Trevett Vice President Mobile Content, NVIDIA President, The Khronos Group Copyright Khronos Group, 2012 - Page 2 Processor

More information

Static Analysis for C++ with Phasar

Static Analysis for C++ with Phasar Static Analysis for C++ with Phasar Philipp Schubert philipp.schubert@upb.de Ben Hermann ben.hermann@upb.de Eric Bodden eric.bodden@upb.de Who are we? Philipp Schubert Chief Developer of PHASAR Teaches

More information

SDAccel Development Environment User Guide

SDAccel Development Environment User Guide SDAccel Development Environment User Guide Features and Development Flows Revision History The following table shows the revision history for this document. Date Version Revision 05/13/2016 2016.1 Added

More information

Standards for Vision Processing and Neural Networks

Standards for Vision Processing and Neural Networks Copyright Khronos Group 2017 - Page 1 Standards for Vision Processing and Neural Networks Radhakrishna Giduthuri, AMD radha.giduthuri@ieee.org Agenda Why we need a standard? Khronos NNEF Khronos OpenVX

More information

Baggy bounds with LLVM

Baggy bounds with LLVM Baggy bounds with LLVM Anton Anastasov Chirantan Ekbote Travis Hance 6.858 Project Final Report 1 Introduction Buffer overflows are a well-known security problem; a simple buffer-overflow bug can often

More information

cuda-on-cl A compiler and runtime for running NVIDIA CUDA C++11 applications on OpenCL 1.2 devices Hugh Perkins (ASAPP)

cuda-on-cl A compiler and runtime for running NVIDIA CUDA C++11 applications on OpenCL 1.2 devices Hugh Perkins (ASAPP) cuda-on-cl A compiler and runtime for running NVIDIA CUDA C++11 applications on OpenCL 1.2 devices Hugh Perkins (ASAPP) Demo: CUDA on Intel HD5500 global void setvalue(float *data, int idx, float value)

More information

Dynamic Dispatch and Duck Typing. L25: Modern Compiler Design

Dynamic Dispatch and Duck Typing. L25: Modern Compiler Design Dynamic Dispatch and Duck Typing L25: Modern Compiler Design Late Binding Static dispatch (e.g. C function calls) are jumps to specific addresses Object-oriented languages decouple method name from method

More information

SIGGRAPH Briefing August 2014

SIGGRAPH Briefing August 2014 Copyright Khronos Group 2014 - Page 1 SIGGRAPH Briefing August 2014 Neil Trevett VP Mobile Ecosystem, NVIDIA President, Khronos Copyright Khronos Group 2014 - Page 2 Significant Khronos API Ecosystem Advances

More information

Boost.Compute. A C++ library for GPU computing. Kyle Lutz

Boost.Compute. A C++ library for GPU computing. Kyle Lutz Boost.Compute A C++ library for GPU computing Kyle Lutz GPUs (NVIDIA, AMD, Intel) Multi-core CPUs (Intel, AMD) STL for Parallel Devices Accelerators (Xeon Phi, Adapteva Epiphany) FPGAs (Altera, Xilinx)

More information

Compiler Tools for HighLevel Parallel Languages

Compiler Tools for HighLevel Parallel Languages Compiler Tools for HighLevel Parallel Languages Paul Keir Codeplay Software Ltd. LEAP Conference May 21st 2013 Presentation Outline Introduction EU Framework 7 Project: LPGPU Offload C++ for PS3 Memory

More information

Getting started. Roel Jordans

Getting started. Roel Jordans Getting started Roel Jordans Goal Translate a program from a high level language into something the processor can execute efficiently So before we start we need to know how this processor executes a program

More information

naïve GPU kernels generation from Fortran source code Dmitry Mikushin

naïve GPU kernels generation from Fortran source code Dmitry Mikushin KernelGen naïve GPU kernels generation from Fortran source code Dmitry Mikushin Contents Motivation and target Assembling our own toolchain: schemes and details Toolchain usecase: sincos example Development

More information

Colin Riddell GPU Compiler Developer Codeplay Visit us at

Colin Riddell GPU Compiler Developer Codeplay Visit us at OpenCL Colin Riddell GPU Compiler Developer Codeplay Visit us at www.codeplay.com 2 nd Floor 45 York Place Edinburgh EH1 3HP United Kingdom Codeplay Overview of OpenCL Codeplay + OpenCL Our technology

More information

Improving Compiler Optimizations using Program Annotations

Improving Compiler Optimizations using Program Annotations Improving Compiler Optimizations using Program Annotations BY NIKO ZARZANI Laurea, Politecnico di Milano, Milan, Italy, 2011 THESIS Submitted as partial fulfillment of the requirements for the degree of

More information

Modern C++ Parallelism from CPU to GPU

Modern C++ Parallelism from CPU to GPU Modern C++ Parallelism from CPU to GPU Simon Brand @TartanLlama Senior Software Engineer, GPGPU Toolchains, Codeplay C++ Russia 2018 2018-04-21 Agenda About me and Codeplay C++17 CPU Parallelism Third-party

More information

Introduction to C++ Introduction to C++ 1

Introduction to C++ Introduction to C++ 1 1 What Is C++? (Mostly) an extension of C to include: Classes Templates Inheritance and Multiple Inheritance Function and Operator Overloading New (and better) Standard Library References and Reference

More information

A Framework for Automatic OpenMP Code Generation

A Framework for Automatic OpenMP Code Generation 1/31 A Framework for Automatic OpenMP Code Generation Raghesh A (CS09M032) Guide: Dr. Shankar Balachandran May 2nd, 2011 Outline 2/31 The Framework An Example Necessary Background Polyhedral Model SCoP

More information

Higher Level Programming Abstractions for FPGAs using OpenCL

Higher Level Programming Abstractions for FPGAs using OpenCL Higher Level Programming Abstractions for FPGAs using OpenCL Desh Singh Supervising Principal Engineer Altera Corporation Toronto Technology Center ! Technology scaling favors programmability CPUs."#/0$*12'$-*

More information

Enabling a Richer Multimedia Experience with GPU Compute. Roberto Mijat Visual Computing Marketing Manager

Enabling a Richer Multimedia Experience with GPU Compute. Roberto Mijat Visual Computing Marketing Manager Enabling a Richer Multimedia Experience with GPU Compute Roberto Mijat Visual Computing Marketing Manager 1 What is GPU Compute Operating System and most application processing continue to reside on the

More information

THE PROGRAMMER S GUIDE TO THE APU GALAXY. Phil Rogers, Corporate Fellow AMD

THE PROGRAMMER S GUIDE TO THE APU GALAXY. Phil Rogers, Corporate Fellow AMD THE PROGRAMMER S GUIDE TO THE APU GALAXY Phil Rogers, Corporate Fellow AMD THE OPPORTUNITY WE ARE SEIZING Make the unprecedented processing capability of the APU as accessible to programmers as the CPU

More information

Press Briefing SIGGRAPH 2015 Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem. Copyright Khronos Group Page 1

Press Briefing SIGGRAPH 2015 Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem. Copyright Khronos Group Page 1 Press Briefing SIGGRAPH 2015 Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem Copyright Khronos Group 2015 - Page 1 Khronos Connects Software to Silicon Open Consortium creating ROYALTY-FREE,

More information

Navigating the Vision API Jungle: Which API Should You Use and Why? Embedded Vision Summit, May 2015

Navigating the Vision API Jungle: Which API Should You Use and Why? Embedded Vision Summit, May 2015 Copyright Khronos Group 2015 - Page 1 Navigating the Vision API Jungle: Which API Should You Use and Why? Embedded Vision Summit, May 2015 Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem

More information

A Brief Introduction to Using LLVM. Nick Sumner

A Brief Introduction to Using LLVM. Nick Sumner A Brief Introduction to Using LLVM Nick Sumner What is LLVM? A compiler? (clang) What is LLVM? A compiler? (clang) A set of formats, libraries, and tools. What is LLVM? A compiler? (clang) A set of formats,

More information

PARALUTION - a Library for Iterative Sparse Methods on CPU and GPU

PARALUTION - a Library for Iterative Sparse Methods on CPU and GPU - a Library for Iterative Sparse Methods on CPU and GPU Dimitar Lukarski Division of Scientific Computing Department of Information Technology Uppsala Programming for Multicore Architectures Research Center

More information

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 Gaming Market Briefing Overview of APIs GDC March 2016 Neil Trevett Khronos President NVIDIA Vice President Developer Ecosystem ntrevett@nvidia.com @neilt3d Copyright Khronos Group 2016 - Page 1 Copyright

More information

Open API Standards for Mobile Graphics, Compute and Vision Processing GTC, March 2014

Open API Standards for Mobile Graphics, Compute and Vision Processing GTC, March 2014 Open API Standards for Mobile Graphics, Compute and Vision Processing GTC, March 2014 Neil Trevett Vice President Mobile Ecosystem, NVIDIA President Khronos Copyright Khronos Group 2014 - Page 1 Khronos

More information

Update on Khronos Open Standard APIs for Vision Processing Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem

Update on Khronos Open Standard APIs for Vision Processing Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem Update on Khronos Open Standard APIs for Vision Processing Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem Copyright Khronos Group 2015 - Page 1 Copyright Khronos Group 2015 - Page

More information

THE NVIDIA DEEP LEARNING ACCELERATOR

THE NVIDIA DEEP LEARNING ACCELERATOR THE NVIDIA DEEP LEARNING ACCELERATOR INTRODUCTION NVDLA NVIDIA Deep Learning Accelerator Developed as part of Xavier NVIDIA s SOC for autonomous driving applications Optimized for Convolutional Neural

More information

More performance options

More performance options More performance options OpenCL, streaming media, and native coding options with INDE April 8, 2014 2014, Intel Corporation. All rights reserved. Intel, the Intel logo, Intel Inside, Intel Xeon, and Intel

More information

OpenCL TM & OpenMP Offload on Sitara TM AM57x Processors

OpenCL TM & OpenMP Offload on Sitara TM AM57x Processors OpenCL TM & OpenMP Offload on Sitara TM AM57x Processors 1 Agenda OpenCL Overview of Platform, Execution and Memory models Mapping these models to AM57x Overview of OpenMP Offload Model Compare and contrast

More information

Accelerating Vision Processing

Accelerating Vision Processing Accelerating Vision Processing Neil Trevett Vice President Mobile Ecosystem at NVIDIA President of Khronos and Chair of the OpenCL Working Group SIGGRAPH, July 2016 Copyright Khronos Group 2016 - Page

More information

Targeting LLVM IR. LLVM IR, code emission, assignment 4

Targeting LLVM IR. LLVM IR, code emission, assignment 4 Targeting LLVM IR LLVM IR, code emission, assignment 4 LLVM Overview Common set of tools & optimizations for compiling many languages to many architectures (x86, ARM, PPC, ASM.js). Integrates AOT & JIT

More information

HKG OpenCL Support by NNVM & TVM. Jammy Zhou - Linaro

HKG OpenCL Support by NNVM & TVM. Jammy Zhou - Linaro HKG18-417 OpenCL Support by NNVM & TVM Jammy Zhou - Linaro Agenda OpenCL Overview OpenCL in NNVM & TVM Current Status OpenCL Introduction Open Computing Language Open standard maintained by Khronos with

More information

HETEROGENEOUS SYSTEM ARCHITECTURE: PLATFORM FOR THE FUTURE

HETEROGENEOUS SYSTEM ARCHITECTURE: PLATFORM FOR THE FUTURE HETEROGENEOUS SYSTEM ARCHITECTURE: PLATFORM FOR THE FUTURE Haibo Xie, Ph.D. Chief HSA Evangelist AMD China OUTLINE: The Challenges with Computing Today Introducing Heterogeneous System Architecture (HSA)

More information

Programming GPUs with C++14 and Just-In-Time Compilation

Programming GPUs with C++14 and Just-In-Time Compilation Programming GPUs with C++14 and Just-In-Time Compilation Michael Haidl, Bastian Hagedorn, and Sergei Gorlatch University of Muenster {m.haidl, b.hagedorn, gorlatch}@uni-muenster.de Abstract. Systems that

More information

HSAIL: PORTABLE COMPILER IR FOR HSA

HSAIL: PORTABLE COMPILER IR FOR HSA HSAIL: PORTABLE COMPILER IR FOR HSA HOT CHIPS TUTORIAL - AUGUST 2013 BEN SANDER AMD SENIOR FELLOW STATE OF GPU COMPUTING GPUs are fast and power efficient : high compute density per-mm and per-watt But:

More information

Standards Update. Copyright Khronos Group Page 1

Standards Update. Copyright Khronos Group Page 1 Standards Update VR/AR, 3D, Web, Vision and Deep Learning Neil Trevett Khronos President NVIDIA VP Developer Ecosystem ntrevett@nvidia.com @neilt3d www.khronos.org Copyright Khronos Group 2017 - Page 1

More information

P4LLVM: An LLVM based P4 Compiler

P4LLVM: An LLVM based P4 Compiler P4LLVM: An LLVM based P4 Compiler Tharun Kumar Dangeti, Venkata Keerthy Soundararajan, Ramakrishna Upadrasta Indian Institute of Technology Hyderabad First P4 European Workshop - P4EU September 24 th,

More information

Khronos Connects Software to Silicon

Khronos Connects Software to Silicon Press Pre-Briefing GDC 2015 Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem All Materials Embargoed Until Tuesday 3 rd March, 12:01AM Pacific Time Copyright Khronos Group 2015 - Page

More information

Copyright Khronos Group, Page 1. OpenCL Overview. February 2010

Copyright Khronos Group, Page 1. OpenCL Overview. February 2010 Copyright Khronos Group, 2011 - Page 1 OpenCL Overview February 2010 Copyright Khronos Group, 2011 - Page 2 Khronos Vision Billions of devices increasing graphics, compute, video, imaging and audio capabilities

More information

Compiler Construction: LLVMlite

Compiler Construction: LLVMlite Compiler Construction: LLVMlite Direct compilation Expressions X86lite Input Output Compile directly from expression language to x86 Syntax-directed compilation scheme Special cases can improve generated

More information

Understanding Undefined Behavior

Understanding Undefined Behavior Session Developer Tools #WWDC17 Understanding Undefined Behavior 407 Fred Riss, Clang Team Ryan Govostes, Security Engineering and Architecture Team Anna Zaks, Program Analysis Team 2017 Apple Inc. All

More information

Taipei Embedded Outreach OpenCL DSP Profile Proposals

Taipei Embedded Outreach OpenCL DSP Profile Proposals Copyright 2018 The Khronos Group Inc. Page 1 Taipei Embedded Outreach OpenCL DSP Profile Proposals Prof. Jenq-Kuen Lee, NTHU Taipei, January 2018 Copyright 2018 The Khronos Group Inc. Page 2 Outline Speaker

More information

Threaded Programming. Lecture 9: Alternatives to OpenMP

Threaded Programming. Lecture 9: Alternatives to OpenMP Threaded Programming Lecture 9: Alternatives to OpenMP What s wrong with OpenMP? OpenMP is designed for programs where you want a fixed number of threads, and you always want the threads to be consuming

More information

Michel Steuwer.

Michel Steuwer. Michel Steuwer http://homepages.inf.ed.ac.uk/msteuwer/ SKELCL: Algorithmic Skeletons for GPUs X i a i b i = reduce (+) 0 (zip ( ) A B) #include #include #include

More information

Lecture 6 More on the LLVM Compiler

Lecture 6 More on the LLVM Compiler Lecture 6 More on the LLVM Compiler Jonathan Burket Special thanks to Deby Katz, Luke Zarko, and Gabe Weisz for their slides Visualizing the LLVM Compiler System C C++ Java Source Code Clang (Front End)

More information

Development of a Translator from LLVM to ACL2

Development of a Translator from LLVM to ACL2 Development of a Translator from LLVM to ACL2 David Hardin, Jennifer Davis, David Greve, and Jedidiah McClurg July 2014 Introduction Research objectives: Reason about machine code generated from high-level

More information

Beyond Hardware IP An overview of Arm development solutions

Beyond Hardware IP An overview of Arm development solutions Beyond Hardware IP An overview of Arm development solutions 2018 Arm Limited Arm Technical Symposia 2018 Advanced first design cost (US$ million) IC design complexity and cost aren t slowing down 542.2

More information

6.S096 Lecture 4 Style and Structure

6.S096 Lecture 4 Style and Structure 6.S096 Lecture 4 Style and Structure Transition from C to C++ Andre Kessler Andre Kessler 6.S096 Lecture 4 Style and Structure 1 / 24 Outline 1 Assignment Recap 2 Headers and multiple files 3 Coding style

More information

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 Open Standards and Open Source Together How Khronos APIs Accelerate Fast and Cool Applications Neil Trevett Khronos President NVIDIA Vice President Mobile Ecosystem Copyright Khronos Group 2015 - Page

More information

Stream Computing using Brook+

Stream Computing using Brook+ Stream Computing using Brook+ School of Electrical Engineering and Computer Science University of Central Florida Slides courtesy of P. Bhaniramka Outline Overview of Brook+ Brook+ Software Architecture

More information

OpenCL Overview Benedict R. Gaster, AMD

OpenCL Overview Benedict R. Gaster, AMD Copyright Khronos Group, 2011 - Page 1 OpenCL Overview Benedict R. Gaster, AMD March 2010 The BIG Idea behind OpenCL OpenCL execution model - Define N-dimensional computation domain - Execute a kernel

More information

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 OpenCL A State of the Union Neil Trevett Khronos President NVIDIA Vice President Developer Ecosystem OpenCL Working Group Chair ntrevett@nvidia.com @neilt3d Vienna, April 2016 Copyright Khronos Group 2016

More information

OpenCL 2.0, OpenCL SYCL & OpenMP 4

OpenCL 2.0, OpenCL SYCL & OpenMP 4 OpenCL 2.0, OpenCL SYCL & OpenMP 4 Open Standards for Heterogeneous Parallel Computing Ronan Keryell AMD Performance & Application Engineering Sunnyvale, California, USA 2014/07/03 High Performance Computing

More information

September 19,

September 19, September 19, 2013 1 Problems with previous examples Changes to the implementation will require recompilation & relinking of clients Extensions will require access to the source code Solutions Combine

More information

Computer Programming: C++

Computer Programming: C++ The Islamic University of Gaza Engineering Faculty Department of Computer Engineering Fall 2017 ECOM 2003 Muath i.alnabris Computer Programming: C++ Experiment #7 Arrays Part II Passing Array to a Function

More information

HSA foundation! Advanced Topics on Heterogeneous System Architectures. Politecnico di Milano! Seminar Room A. Alario! 23 November, 2015!

HSA foundation! Advanced Topics on Heterogeneous System Architectures. Politecnico di Milano! Seminar Room A. Alario! 23 November, 2015! Advanced Topics on Heterogeneous System Architectures HSA foundation! Politecnico di Milano! Seminar Room A. Alario! 23 November, 2015! Antonio R. Miele! Marco D. Santambrogio! Politecnico di Milano! 2

More information

HSA Foundation! Advanced Topics on Heterogeneous System Architectures. Politecnico di Milano! Seminar Room (Bld 20)! 15 December, 2017!

HSA Foundation! Advanced Topics on Heterogeneous System Architectures. Politecnico di Milano! Seminar Room (Bld 20)! 15 December, 2017! Advanced Topics on Heterogeneous System Architectures HSA Foundation! Politecnico di Milano! Seminar Room (Bld 20)! 15 December, 2017! Antonio R. Miele! Marco D. Santambrogio! Politecnico di Milano! 2

More information

4/1/15 LLVM AND SSA. Low-Level Virtual Machine (LLVM) LLVM Compiler Infrastructure. LL: A Subset of LLVM. Basic Blocks

4/1/15 LLVM AND SSA. Low-Level Virtual Machine (LLVM) LLVM Compiler Infrastructure. LL: A Subset of LLVM. Basic Blocks 4//5 Low-Level Virtual Machine (LLVM) LLVM AND SSA Slides adapted from those prepared by Steve Zdancewic at Penn Open-Source Compiler Infrastructure see llvm.org for full documntation Created by Chris

More information

Alias Analysis in LLVM

Alias Analysis in LLVM Alias Analysis in LLVM by Sheng-Hsiu Lin Presented to the Graduate and Research Committee of Lehigh University in Candidacy for the Degree of Master of Science in Computer Science Lehigh University May

More information

WebGL Meetup GDC Copyright Khronos Group, Page 1

WebGL Meetup GDC Copyright Khronos Group, Page 1 WebGL Meetup GDC 2012 Copyright Khronos Group, 2012 - Page 1 Copyright Khronos Group, 2012 - Page 2 Khronos API Ecosystem Trends Neil Trevett Vice President Mobile Content, NVIDIA President, The Khronos

More information

Deep Learning Frameworks. COSC 7336: Advanced Natural Language Processing Fall 2017

Deep Learning Frameworks. COSC 7336: Advanced Natural Language Processing Fall 2017 Deep Learning Frameworks COSC 7336: Advanced Natural Language Processing Fall 2017 Today s lecture Deep learning software overview TensorFlow Keras Practical Graphical Processing Unit (GPU) From graphical

More information

(5-1) Object-Oriented Programming (OOP) and C++ Instructor - Andrew S. O Fallon CptS 122 (February 4, 2019) Washington State University

(5-1) Object-Oriented Programming (OOP) and C++ Instructor - Andrew S. O Fallon CptS 122 (February 4, 2019) Washington State University (5-1) Object-Oriented Programming (OOP) and C++ Instructor - Andrew S. O Fallon CptS 122 (February 4, 2019) Washington State University Key Concepts 2 Object-Oriented Design Object-Oriented Programming

More information

Copyright Khronos Group Page 1

Copyright Khronos Group Page 1 Update on Khronos Standards for Vision and Machine Learning December 2017 Neil Trevett Khronos President NVIDIA VP Developer Ecosystem ntrevett@nvidia.com @neilt3d www.khronos.org Copyright Khronos Group

More information

INTRODUCTION TO OPENCL TM A Beginner s Tutorial. Udeepta Bordoloi AMD

INTRODUCTION TO OPENCL TM A Beginner s Tutorial. Udeepta Bordoloi AMD INTRODUCTION TO OPENCL TM A Beginner s Tutorial Udeepta Bordoloi AMD IT S A HETEROGENEOUS WORLD Heterogeneous computing The new normal CPU Many CPU s 2, 4, 8, Very many GPU processing elements 100 s Different

More information

WebGL, WebCL and OpenCL

WebGL, WebCL and OpenCL Copyright Khronos Group, 2011 - Page 1 WebGL, WebCL and OpenCL Neil Trevett Vice President Mobile Content, NVIDIA President, The Khronos Group Copyright Khronos Group, 2011 - Page 2 Processor Parallelism

More information

Silicon Acceleration APIs

Silicon Acceleration APIs Copyright Khronos Group 2016 - Page 1 Silicon Acceleration APIs Embedded Technology 2016, Yokohama Neil Trevett Vice President Developer Ecosystem, NVIDIA President, Khronos ntrevett@nvidia.com @neilt3d

More information

Jose Aliaga (Universitat Jaume I, Castellon, Spain), Ruyman Reyes, Mehdi Goli (Codeplay Software) 2017 Codeplay Software Ltd.

Jose Aliaga (Universitat Jaume I, Castellon, Spain), Ruyman Reyes, Mehdi Goli (Codeplay Software) 2017 Codeplay Software Ltd. SYCL-BLAS: LeveragingSYCL-BLAS Expression Trees for Linear Algebra Jose Aliaga (Universitat Jaume I, Castellon, Spain), Ruyman Reyes, Mehdi Goli (Codeplay Software) 1 About me... Phd in Compilers and Parallel

More information